From October 17th to October 19th, the 2024 World Intelligent Connected Vehicle Conference, jointly sponsored by the Ministry of Industry and Information Technology, the Ministry of Transport, and the Beijing Municipal People's Government, was held in Beijing.
Among them, vehicle-road-cloud integration, as one of the key technologies of intelligent networked vehicles, has become a hot topic of discussion at the conference. As a leading provider of intelligent networked vehicle solutions in China, Zhu Lei, the founder of Mushroom Car Link, was interviewed by a reporter from China Automotive News during the conference, in-depth discussion of the key issues faced in the development of vehicle-road-cloud integration, and looked forward to the future development direction.
The importance of "network" in vehicle-road-cloud integration
At present, technologies such as vehicle-road-cloud integration are in a critical period of testing and verification to large-scale application. The core of vehicle-road-cloud integration is to realize the deep integration of vehicles, road infrastructure and cloud data. Through this integration, the operational efficiency and safety of intelligent networked vehicles can be effectively improved, and more accurate decision-making support can be provided for urban traffic management. However, there are still a number of challenges that need to be addressed to achieve this goal.
When it comes to vehicle-road-cloud integration, Zhu Lei particularly emphasized the importance of the network. With the development, the connotation of vehicle-road-cloud integration is gradually evolving into a "vehicle-road-cloud network", which is essentially a set of "synaesthesia computing" network, which provides real-time data services for all smart devices by realizing the real-time digitization of the physical world, so that smart devices can interact with the real world in real time.
The core of the future development of vehicle-road-cloud integration technology lies in building a comprehensive real-time digital twin system, which can not only achieve accurate perception of the physical world, but also provide real-time data services, so that smart devices can seamlessly interact with the real world. At present, this technology is mainly used in intelligent driving, forming a so-called "vehicle-to-cloud" network, but its potential is much more than that.
However, for the perception level, comprehensive real-time perception is particularly important. The real-time requirement for the time of the libre is getting shorter and shorter, which puts forward higher requirements for the instantaneity of the network, which requires the vehicle-road cloud network to provide full real-time information of the traffic environment. Among them, the efficiency problems faced by them need to be solved at the system level through the vehicle-road-cloud network.
"In the process of building a vehicle-to-road cloud network, there are still many pain points in the industry, such as the low data quality of a single node in the network." Zhu Lei said that in the future, the realization of large-scale real-time digital twins needs to do two things, one is that the data quality of a single node (base station) needs to be high enough, and the core of ensuring data quality is not only hardware equipment, but systems and algorithms. Vehicle-road-cloud integration is not simply a stack of hardware and equipment, but requires a set of operating systems to integrate and drive hardware such as cameras, lidars, and communication computing, and process data through the AI algorithm model in the operating system. The second is to have an in-depth understanding of the real needs of car companies and truly connect with car companies. The data of vehicle-road-cloud integration is not a communication indicator in the usual sense, but a set of perception indicators that can be recognized by the vehicle, such as the completeness of recognition, accuracy, feedback time, etc., which has a high degree of similarity with the perception system of autonomous driving.
"Only by doing these two things can we realize the data on the car on a large scale, and the integration of vehicles, roads and clouds can be truly used and generate value." Zhu Lei revealed to reporters that at present, Mushroom Auto Union is cooperating with a number of vehicle companies to jointly promote the commercialization of vehicle-road-cloud integration technology.
In addition, the security of intelligent networked vehicles is particularly important in the context of the rapid development of technology. In this regard, the roadside transportation infrastructure is integrated into the system of intelligent networked vehicles, and the vehicle no longer only relies on its own sensors to judge the surrounding environment, but can obtain more extensive and comprehensive information with the help of roadside infrastructure, and even some more accurate data, so as to improve traffic safety.
The integration of vehicles, roads and clouds enables the high-quality development of the industry
Zhu Lei said that vehicle-road-cloud integration is a typical scenario of new quality productivity, with a high degree of technology integration and intelligence, collaboration with multiple subdivided science and technology fields, and real-time interaction with a variety of intelligent devices, bringing value to urban and traffic management. With the further development of technology, the vehicle-road-cloud network will be extended to the entire city, and it is expected that a digital twin system covering the country and even the world will eventually be formed. Thanks to the rapid advances in artificial intelligence, computing power, and big data over the past decade, this technology is able to perceive the environment in all directions and with high accuracy, and process and transmit data in a very short time.
Zhu Lei said that the development of the vehicle-road cloud network will mainly bring three aspects of value, one is to solve the traffic problem, large-scale access to the vehicle-road cloud network can solve the problem of traffic safety, efficiency, and game decision-making, such as "zero accidents" in terms of safety, efficiency improvement, and better game decision-making. The second is the value of data, the connected sensors and agents form a data network, which can obtain the global digital information of the physical world, improve the efficiency of data acquisition, and feed back a variety of model training. The third is to build an intelligent society, help the digital intelligent society enter the integrated system, and intelligent devices can access the corresponding information to achieve information sharing and collaboration.
"Based on different communication methods, two sets of relatively mature solutions, namely C-V2X and 5G-A, have been formed in each city, and each city can flexibly deploy them according to its own infrastructure conditions, and its commercialization path has gradually become clear." Zhu Lei said that the business model of the vehicle-road-cloud industry is similar to that of the communications industry, where vehicle-road-cloud network operators build network infrastructure and benefit from providing real-time digital twin data services to end vehicles and users. At the same time, the vehicle-road cloud network can provide traffic management and dispatching services for cities and vehicle operating units, and create benefits by optimizing traffic flow and improving transportation efficiency. In addition to providing services to vehicles, vehicle-to-road cloud networks can also be extended to a variety of agents, which will also give rise to a variety of application scenarios and business models that are currently unimaginable.
Zhu Lei said that with the continuous maturity of technology, the services of the vehicle-road cloud network will no longer be limited to the automotive field, whether it is a motor vehicle, an aircraft, a robot, a robot dog, or even an AI mobile phone, AIPC and other intelligent terminals can access this network. These smart devices will be based on data information for real-time decision-making and precise planning, enabling more efficient collaboration. Based on this, the integration of vehicles, roads and clouds will also play an important role in the future competition between major countries in science and technology.